Prediction of Heart Disease using Naïve Bayes Technique of Data Mining

Abstract

Coronary illness is responsible for deaths in all age groups and is common among males and females. An excellent answer for this issue is to have the option to predict what a patient's health status will in future so the specialists can begin treatment much sooner which will yield better outcomes. Data mining plays most significant role in area of investigation by means of the objective to finding essential data from massive amount of information. Currently, data mining strategies and tools are utilized by researchers in the field of healthcare, especially for prediction of sickness. Data mining methodology affords improvement approach to interchange huge data into beneficial information for attaining selection. In utilising data mining patterns they desires considerably fewer amount of funding intended for the forecasting the ailment alongside better accurate and precision. Moreover, analysis of study paper depicts the estimation of coronary illness in clinical field by utilizing data mining. Various popular data mining algorithm on the dataset of 13 attributes is applied to forecast the coronary ailment at initial stage. The dataset is collected from UCI machine learning repository and analysed with various parameters like Accuracy, Recall, Precision, F-measure, ROC area and Kappa statistics. Experimental results show that the Naïve bayes algorithm is always becomes the best-performing data mining method which accomplishes an accuracy of 86.716% in coronary illness prediction.